CN103262124B - For the imaging system to region of interest domain imaging - Google Patents

For the imaging system to region of interest domain imaging Download PDF

Info

Publication number
CN103262124B
CN103262124B CN201180060794.5A CN201180060794A CN103262124B CN 103262124 B CN103262124 B CN 103262124B CN 201180060794 A CN201180060794 A CN 201180060794A CN 103262124 B CN103262124 B CN 103262124B
Authority
CN
China
Prior art keywords
component
value
interest
pad value
area
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201180060794.5A
Other languages
Chinese (zh)
Other versions
CN103262124A (en
Inventor
R·普罗克绍
B·J·布伦德尔
D·T·克勒
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koninklijke Philips NV
Original Assignee
Koninklijke Philips Electronics NV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips Electronics NV filed Critical Koninklijke Philips Electronics NV
Publication of CN103262124A publication Critical patent/CN103262124A/en
Application granted granted Critical
Publication of CN103262124B publication Critical patent/CN103262124B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/002D [Two Dimensional] image generation
    • G06T11/003Reconstruction from projections, e.g. tomography
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/002D [Two Dimensional] image generation
    • G06T11/003Reconstruction from projections, e.g. tomography
    • G06T11/006Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2211/00Image generation
    • G06T2211/40Computed tomography
    • G06T2211/408Dual energy
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2211/00Image generation
    • G06T2211/40Computed tomography
    • G06T2211/424Iterative

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Algebra (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

The noise of the probe value gathered by imaging system (30) depends on the contribution of different component being imaged in area-of-interest, radiation (4) through described different component to produce collection probe value respectively.This dependence is considered when rebuilding the image of described area-of-interest iteratively, wherein, determine the first component pad value corresponding with the element of the first component in described area-of-interest, and the second component pad value corresponding with the element of the second component in described area-of-interest, wherein, determine noise figure from described first component pad value and described second component pad value, and wherein, use described noise figure to update described image.Consider the described dependence of noise gathering probe value on described different component improve described in the quality of image rebuild iteratively.

Description

For the imaging system to region of interest domain imaging
Technical field
The present invention relates to a kind of imaging system, a kind of formation method with one for area-of-interest is become The tomography computer program of picture.
Background technology
US 2003/0156684 A1 discloses a kind of image reconstructor device, and it is for having from multiple The multifarious transmission measurement of energy statistically rebuilds image.Described device includes: be used for providing multiple tool There is the equipment of the multifarious transmission measurement of energy, and for the equipment by measuring described in algorithm process, Based on statistical model, described algorithm considers that described energy multiformity has reduction noise to obtain at least one Final constitutional diagram picture.
Article " the Ordered subsets algorithms for of H.Erdogan and J.A.Fessler Transmission tomography ", Physics in Medicine and Biology, 44 (11), November 1999, discloses a kind of for from the spy by using computed tomography systems to gather Measured value rebuilds the iterative reconstruction approach of image iteratively.In iterative step, performed by present image Forward projection, to generate analog detection value, determines between described analog detection value and the probe value gathered Difference, and use noise model to calculate renewal, the most backward throwing of described renewal from described difference Shadow updates described present image.Perform these iterative steps some, until meeting end condition.To the greatest extent Pipe is by using noise model to calculate described renewal, but still reduces the quality of described reconstruction image.
Summary of the invention
The present invention aims at a kind of imaging system of offer, a kind of formation method and one for sense The tomography computer program of region of interest domain imaging, wherein it is possible to improve the matter of the image rebuild iteratively Amount.
In the first aspect of the invention, it is proposed that a kind of for the imaging to region of interest domain imaging System, wherein, described imaging system includes:
-probe value provides unit, and it is for providing the collection probe value of described area-of-interest, wherein, Described area-of-interest includes at least two component, and wherein, described probe value is own by relative to each other Move and launch the radiation source and described area-of-interest radiated and by detecting described probe value collection, The instruction of described probe value passes institute when described radiation source and described area-of-interest are moved relative to State the radiation after area-of-interest;
-reconstruction unit, it is for rebuilding described area-of-interest iteratively by the some iterative steps of execution Final image, in described iterative step based on described collection probe value and indicate described probe value make an uproar The noise figure of sound updates intermediate image, and wherein, described reconstruction unit is suitable to:
-determine the first component pad value and second component pad value, wherein, described first component pad value Corresponding with the element of the first component in described area-of-interest, and wherein, described second component Pad value is corresponding with the element of the second component in described area-of-interest;
-determine described noise figure from described first component pad value and described second component pad value.
The described noise gathering probe value may rely on the contribution of different component, and described radiation is through institute State different component and produce collection probe value respectively.Determining that the first component pad value and second component decline In the case of depreciation, and determining from described first component pad value and described second component pad value In the case of the noise figure updating described respective intermediate image, described sense can be rebuild iteratively Considering this dependence during the image in interest region, wherein, described first component pad value is with described The element of the first component in area-of-interest is corresponding, and wherein, described second component pad value Corresponding with the element of the second component in described area-of-interest.So consider on described different component Gather the described dependence of the noise of probe value, improve the quality of described iterative approximation image, institute State to radiate and produce collection probe value respectively through described different component.
Described area-of-interest can include plural component, and described reconstruction unit can be fitted In determining two or more component pad value, wherein, the component more than the two in this case Pad value determines described noise figure.Described different component can be different such as bone and the basis of soft tissue Material, and/or they can relate to different physical effects, as photoelectric effect, Compton effect, K side effect etc..
Being preferably polychromatic radiation by the described radiation of described radiation emission, described polychromatic radiation is relatively Low-yield decayed more consumingly by the described component in described area-of-interest than higher-energy.This leads Causing the beam hardening of so-called radiation, described beam hardening limits the quality of described collection probe value, And thus, limit the quality by the image using described collection probe value to rebuild iteratively.Cause This preferably, described reconstruction unit be suitable to based on determined by the first component pad value and determined by the Two component pad values are for gathering probe value described in beam hardening correction.Such for beam hardening Revise the quality of the image rebuild iteratively described in further improving.
The described of described noise figure determines and can perform before described iterative approximation starts, or in institute During stating iterative approximation one, several or all iterative steps perform.Described collection probe value Described beam hardening correction can also perform before described iterative approximation starts, or described repeatedly Generation rebuild during one, several or all iterative steps perform.
Described probe value provides unit can be the memory element having stored and having gathered probe value, or Described probe value provides unit can be probe value collecting unit, such as X ray computer tomography system System, x-ray C-arm system, such as positron emission tomography system or single photon emission calculate The collecting unit of the nuclear imaging system etc. of machine tomography.
Described reconstruction unit may be adapted to perform in iterative step: by the forward direction of described intermediate image Project to generate the artificial line integration of absorption value;From the line gathering the absorption value that probe value determines respectively Subtracting each other of the described artificial line integration of integration and absorption value;By using described noise figure to calculate from difference Update and calculate the rear orientation projection of described renewal, to update described intermediate image, wherein, if performing Do until meeting end condition in these iterative steps, specifically, until the described simulation of absorption value Difference between line integral and the described line integral of absorption value determined from described collection probe value or work as The front difference between intermediate image and intermediate image before is less than predefined threshold value.Preferably, The acquired projections data of beam hardening correction are used in each iterative step.
Described noise figure is preferably the variance of described collection probe value.This allows updating described centre Described noise is effectively considered during image.
Preferably, according to being determined by the described energy spectrum of the described radiation of described radiation emission Noise figure.Determine noise figure according to by the energy spectrum of the described radiation of described radiation emission, improve The quality of described noise figure, and therefore improve the described noise when rebuilding described image iteratively Consider.This then further improve described in the quality of image rebuild iteratively.
It is further preferred that described reconstruction unit is suitable to determine one accordingly for each collection probe value The first component pad value and corresponding second component pad value between ratio, and be suitable to based on institute State ratio and determine described noise figure.If gathering probe value, the first component pad value and second component to decline Depreciation belongs to identical ray path, then they are the most corresponding.According to corresponding first component Ratio between pad value and corresponding second component pad value determines described noise figure, it is allowed to one Relatively simple method determines high-quality noise figure.
Preferably, described reconstruction unit is adapted to determine that described first component pad value and described second group Divide pad value, pass through:
-determine the first component Parts and second component part, wherein, described first component from intermediate image Part shows the element of described first component in described area-of-interest, and wherein, described second Component Parts shows the element of the described second component in described area-of-interest;
-determine that described first component decays by simulation by the forward projection of described first component Parts Value, and determine that described second component declines by simulation by the forward projection of described second component part Depreciation.It is further preferred that described first component Parts forms the first constitutional diagram picture, and described second Component Parts forms second component image, and wherein, described reconstruction unit is suitable to simulation by described first The forward projection of constitutional diagram picture determines described first component pad value, and is suitable to by simulation by institute The forward projection stating second component image determines described second component pad value.Described intermediate image includes Such as voxel or the pictorial element of pixel.Single image element can include single component Parts or single figure Pixel element can include the first component Parts and second component part.If such as, pictorial element has The image value of 500Hounsfield unit (HU), and if oneself know and deposit in described area-of-interest At the first component corresponding to 0HU and the second component corresponding to 1000HU, then it is believed that institute State described first component and the described second component of 50% that pictorial element represents 50%.
In an embodiment, described first component Parts can be the highly attenuating part of described intermediate image also And described second component part can be the low attenuation portions of described intermediate image, wherein, described height declines Subtract the element in described area-of-interest shown partially with absorption bigger to described radiation, and wherein, Described low attenuation portions illustrates the element in described area-of-interest with absorption less to described radiation. Described highly attenuating part and described low attenuation portions are permissible, such as, determined by threshold process, wherein, It is considered as the image value of low attenuation portions with the image value corresponding less than the radiation absorption of described threshold value, And it is considered as wherein, highly attenuating portion with the image value corresponding more than the radiation absorption of described threshold value The image value divided.This allows relatively simply to determine that the first component declines by threshold process and forward projection Depreciation and second component pad value.
It is also preferred that described probe value provides unit to be adapted to provide for the collection probe value of Energy Dependence, Wherein, described reconstruction unit is suitable to by the collection probe value application foundation component to described Energy Dependence Decomposition method determines described first component pad value and described second component pad value.Preferably implementing In example, the probe value of described Energy Dependence is corresponding with multiple energy branch mailbox (energy bin), wherein, Described reconstruction unit is suitable to be described as the first contribution and the group of the second contribution by use by gathering probe value The simultaneous equations of the probe value for the plurality of Energy Dependence are asked by the model gathering probe value closed Solving, thus apply described infrastructure component to decompose, described first contribution depends on is drawn by described first component The decay of the first Energy Dependence risen, described second contribution depends on the caused by described second component The decay of two Energy Dependences.
Described area-of-interest has the element that relatively large radiation absorbs and can be defined as such as having big There is in the absorption of predefined threshold value, and described area-of-interest the unit of less radiation absorption Element can be defined as the absorption less than predefined threshold value.Such as, described in definable in advance The threshold value of definition so that in the case of the part of human or animal occurs in described area-of-interest, The element in described area-of-interest with less absorption is corresponding with soft tissue, and in described sense Interest region has the element of bigger absorption and the such as bone that possibly be present in described area-of-interest, The contrast agent of metal or such as iodine is corresponding.The most described first component pad value preferably with such as bone Element is corresponding, and the most described second component pad value is preferably relative with such as soft tissue element Should.
In the case of determining two or more component pad value, every kind of component pad value is interested with described Region has the component to described radiation specific absorption spectrum in certain absorption region corresponding, Wherein, in this case it is preferable to determine described noise figure according to described several component pad values, Such as, according to highly attenuating probe value, middle decay probe value and low decay probe value, wherein, described Highly attenuating probe value with described area-of-interest has compared with large radiation absorb element corresponding, institute State middle decay probe value with described area-of-interest has middle radiation absorption element corresponding, Further, described low decay probe value and described area-of-interest have less radiation absorption element Corresponding.In this case, also preferably penetrate according to the execution of described several component pad value Bundle hardening.The quality of the image rebuild iteratively described in so can improving further.
In still yet another aspect of the present, it is proposed that a kind of for the imaging side to region of interest domain imaging Method, wherein, described formation method includes:
-the collection probe value of described area-of-interest is provided, wherein, described area-of-interest include to Few two components, wherein, described probe value is own by being moved relative to launch the radiation source of radiation With described area-of-interest and by detecting described probe value collection, the instruction of described probe value is when described The spoke after described area-of-interest is passed when radiation source and described area-of-interest are moved relative to Penetrate;
-rebuild the final image of described area-of-interest iteratively, in institute by performing some iterative steps Stating in iterative step, noise figure based on described collection probe value and the described probe value noise of instruction updates Intermediate image;Wherein it is determined that the first component pad value and second component pad value, wherein, described One component pad value is corresponding with the element of the first component in described area-of-interest, and wherein, Described second component pad value is corresponding with the element of the second component in described area-of-interest;Wherein, Described noise figure is determined from described first component pad value and described second component pad value.
In still yet another aspect of the present, it is proposed that a kind of based on the imaging to region of interest domain imaging Calculation machine program, wherein, described tomography computer program includes code modules, when described computer When program is run on the computer controlling described imaging system, described code modules makes such as right Require that the imaging system described in 1 performs the step of formation method as claimed in claim 10.
It should be understood that imaging system as claimed in claim 1, imaging as claimed in claim 10 Method has similar and/or identical preferred embodiment with imaging device as claimed in claim 11, as Defined in dependent claims.
It should be understood that the preferred embodiments of the present invention can also be with respective independent claims The combination in any of dependent claims.
Accompanying drawing explanation
In the accompanying drawings:
Fig. 1 diagrammatically and exemplarily illustrates the reality for the imaging system to region of interest domain imaging Execute example;
Flow chart shown in Fig. 2 exemplarily illustrates for the formation method to region of interest domain imaging Embodiment;
Flow chart shown in Fig. 3 exemplarily illustrates some steps of iterative step.
Detailed description of the invention
Fig. 1 diagrammatically and exemplarily illustrates for the imaging system to region of interest domain imaging, institute Stating imaging system is computed tomography systems 30.Described computed tomography systems includes can Along the frame 1 being parallel to the rotary shaft R rotation that z direction extends.Radiation source 2, in the present embodiment It is X-ray tube, is arranged in described frame 1.Described radiation source 2 is provided with collimator 3, at this In embodiment, described collimator 3 forms cone from the described radiation 4 generated by described radiation source 2 Radiant flux.Described radiation is through the area-of-interest of the part including object or object.Described object It is, such as, people, animal or technical object.Described area-of-interest is positioned in inspection area 5, Described inspection area 5 is cylindrical in this embodiment.Described radiant flux is through described test zone After territory 5, incide on the detecting devices 6 including two-dimensional detecting surface.Described detecting devices 6 is pacified It is contained in described frame 1.
Described computed tomography systems includes two motors 7,8.Described motor 7 is preferably with perseverance But fixed adjustable angular velocity drives described frame 1.There is provided described motor 8 for by being parallel to The direction stating rotary shaft R or described z-axis shifts described object (such as, in described inspection area 5 In the people that is arranged on patient table) shift described area-of-interest.These motors 7,8 are by (such as) Control unit 9 controls so that described radiation source 3 and described inspection area 5 and therefore in described inspection Described area-of-interest in region 5 is moved relative to along helical orbit.However, it is also possible to It is to perform described relative movement along other tracks.Such as, it is also possible to, it is described right not move As or described inspection area 5, and only rotate described radiation source, e.g., described radiation source is right relative to described As moving with described inspection area 5 annularly track.Additionally, in another embodiment, described collimation Device 3 may be adapted to be formed another beam shape, particularly forms fan beam, and described detection sets Standby 6 can include having corresponding with other beam shapes described (particularly corresponding with fan beam ) searching surface.
In described radiation source 2 and described inspection area 5 and the relative shifting of the most described area-of-interest Dynamic period, described detecting devices 6 is according on the described searching surface inciding described detecting devices 6 Radiation generates and gathers probe value.Therefore, described radiation source 2, for moving relative to described area-of-interest The element of dynamic described radiation source 2, specifically, described motor 7,8 and described frame 1, and described Detecting devices 6 is formed for providing the probe value gathering probe value of described area-of-interest to provide unit 31。
There is provided described collection probe value to rebuild described area-of-interest iteratively for reconstruction unit 10 Whole image.Described reconstruction unit 10 is also preferably controlled by described control unit 9.Described reconstruction unit 10 are suitable to include described object by perform that some iterative steps rebuild described area-of-interest iteratively The final image of at least two component, based on described collection probe value with refer in described iterative step Show that the noise figure of described collection probe value noise is to update intermediate image.Described noise figure is from also by described The first component pad value and second component pad value that reconstruction unit 10 determines determine.Described first component Pad value is corresponding with the element of the first component in described area-of-interest, and described second component Pad value is corresponding with the element of the second component in described area-of-interest.In this embodiment, institute Stating the first component pad value is highly attenuating probe value, described highly attenuating probe value and described area-of-interest In to have the element of the described bigger absorption of radiation 4 corresponding, and described second component pad value declines with low Subtract probe value corresponding, described low decay probe value and described area-of-interest have less absorption Element is corresponding.Additionally, in this embodiment, described noise figure is the variance of described collection probe value, And described reconstruction unit 10 be suitable to based on determined by the first component pad value and determined by second group Divide pad value for gathering probe value described in beam hardening correction.Described reconstruction unit 10 is the suitableeest In more than according to described first component pad value and described second component pad value, and according to by described The energy spectrum of the described radiation 4 that radiation source 2 is launched is to determine described noise figure.
In this embodiment, described reconstruction unit 10 is suitable to perform in iterative step: in described Between image forward projection with generate absorption value artificial line integration;Determine from collection probe value respectively The line integral of absorption value and the subtracting each other of artificial line integration of absorption value;By use described noise figure from Difference calculates renewal and calculates the rear orientation projection of described renewal, to update described intermediate image, wherein, Perform these iterative steps some until meeting end condition, specifically, until absorption value is described Difference between artificial line integration and the described line integral of absorption value determined from described collection probe value or Difference between intermediate image and intermediate image before that person is current is less than predefined threshold value.? Just, described intermediate image it may be that such as, has null value as the image of image value.Described heavy Build unit 10 be more preferably suitable in iterative step at least one described perform beam hardening correction repair The most described collection probe value, wherein, the collection probe value revised is used in respective iterative step also And the iterative step after being also used in updates described intermediate image.
Described reconstruction unit 10 is preferably adapted to determine described first component pad value and described second component Pad value, by i) determining highly attenuating part and low attenuation portions, wherein, institute from described intermediate image State the element in highly attenuating described area-of-interest shown partially with absorption bigger to described radiation, and And wherein, described low attenuation portions illustrates have absorption less to described radiation in described area-of-interest Element;And ii) determine described first group by simulation through the forward projection of described highly attenuating part Divide pad value, and determine described second group by simulating the forward projection through described low attenuation portions Divide pad value.In this embodiment, described area-of-interest has the bigger absorption of described radiation 4 Element be defined as in the absorption bigger than predefined threshold value, and described area-of-interest The element of the relatively low absorption with described radiation 4 is defined as the suction less than predefined threshold value Receive.In this example, described threshold value is selected so that described area-of-interest has more than described threshold The element of the absorption of value represents and possibly be present at bone, metal and/or the contrast agent in described area-of-interest, And described area-of-interest has and represents soft tissue less than the element of the absorption of described threshold value.Specifically For, determined by threshold process and form the highly attenuating part of highly attenuating image and form low decay pattern picture Low attenuation portions, wherein, if there is described interested in if, described highly attenuating image shows Going out bone, metal and/or contrast agent, described low decay pattern picture illustrates soft tissue.Then can be by simulation Forward projection through described highly attenuating image determines described first component pad value, and can lead to Cross the simulation forward projection through described low decay pattern picture to determine described second component pad value.Therefore Described first component pad value is preferably corresponding with (such as) bone, metal and/or contrast agent element, And the most described second component pad value is preferably corresponding with (such as) soft tissue element.
Described reconstruction unit 10 may also suitably be and determines the first component pad value and second component by other method Pad value, particularly, the collection providing unit 31 to be adapted to provide for Energy Dependence at described probe value detects In the case of value.For providing the probe value of Energy Dependence, described probe value provides the described radiation of unit Source is preferably polychromatic radiation source and described probe value provides the detecting devices of unit to be preferably energy Resolved detection equipment.
Energy-resolved detection equipment counts described incident photon according to (such as) in particular energy branch mailbox The principle work of the signal of energy photon number each with output display.Such as, Llopart, X. et al. Article " First test measurements of a 64k pixel readout chip wordkin in a single Photon counting mode ", Nuclear Instruments and Methods in Physics Research, Section A, Volume 509, Issue 1-3, Pages 157-163 (2003), and Llopart, X. Et al. article " Medipix2:A 64-k pixel readout chip with 55mum square elements Working in a single photon counting mode ", IEEE Transactions on Nuclear Science, Volume 49, Issue 5, Pages 2279-2283 (2000) describe such energy Resolved detection equipment, passes through to quote at this to be incorporated into.
Described reconstruction unit 10 may be adapted to determine described first component by application foundation decomposition of components method Pad value and described second component pad value, described basic components decomposition method be in this embodiment for The basic material decomposition method of described Energy Dependence collection probe value, wherein, first foundation material such as bone, Metal or contrast agent include the bigger absorption to described radiation, and wherein, the second basic material is the softest Tissue includes the less absorption to described radiation.Such as, the probe value of described Energy Dependence can be with multiple Energy branch mailbox is corresponding, and wherein, described reconstruction unit 10 may be adapted to by using probe value description It is the model of the described probe value of the combination of the first contribution and the second contribution, to for the plurality of energy The simultaneous equations of the probe value relied on solve, thus apply described basic components to decompose, described first tribute Offering the decay being depended on the first Energy Dependence by described first foundation material, described second contribution is by described Second basic material depends on the decay of the second Energy Dependence.Specifically, described reconstruction unit 10 can It is suitable for solving following simultaneous equations:
Wherein, SiRepresent the probe value of i-th energy branch mailbox,Represent that radiation will pass described imaging Spectrum before object, δi(E) representing the spectral sensitivity of i-th energy branch mailbox, N represents basic material Number, be such as two for bone and soft tissue N, AjRepresent the linear attenuation of material j, and fj(E) Represent the absorption of the Energy Dependence of material j.
In the case of the number of the described energy branch mailbox number at least equal to described basic material, permissible Described simultaneous equations are solved by known numeric value method, it is known that described amountδiAnd f (E)j(E), And the result solving described simultaneous equations is described linear attenuation Aj.Radiation spectrumWith described light Spectral sensitivity δi(E) there is the characteristic of described imaging system and learn from such as corresponding measurement.State all Bone and soft tissue be modeled material as described in spectral absorption fj(E) also learn from measurement and/or from literary composition Offer and learn.If such as, the number of basic material is two, wherein, the e.g. first foundation of bone Second basic material of material Phase Proportion soft tissue in this way has bigger absorption, described linear attenuation A1 First component pad value determined by expression and described linear attenuation A2Second component determined by expression Pad value.
Although the most preferably determining two kinds of probe values, such as, the first component pad value With second component pad value, but described reconstruction unit 10 can also be adapted to determine that two or more or two classes Above probe value.For example, it may be determined that some different classes of probe values, wherein, each classification Corresponding with the element in described area-of-interest, described element has in certain absorption region institute State the specific absorption spectrum of radiation 4.For example, it may be determined that the first component pad value, second component More than pad value and the 3rd component pad value, or three different types of probe value.Specifically, By using equation (1) to may determine that N kind probe value, wherein, every in the probe value of these kinds One is all corresponding with a certain basic material.For example, it is possible to for soft tissue, bone, metal, such as iodine Contrast agent or other materials determine probe value.
Described reconstruction unit 10 be preferably incorporated in a) the first component pad value, second component pad value, The energy spectrum of the radiation gone out by described radiation emission, and b) gather accordingly between the variance of probe value Distribution.This distribution is such as stored in described reconstruction unit 10 as form or function.By to often Individual collection probe value uses described distribution, can distribute variance.Described distribution can be by such as calibrating survey Amount generates, wherein, for the described different known spectra launching radiation with cause in described ray point Described first component of other collection probe value and the different known ratio of described second component, measure institute State variance.It is likely to by analogue measurement or by using (such as) Poisson noise model to calculate institute State distribution to determine described distribution.Preferably, corresponding first and second components of described basic of distribution decline The ratio of depreciation, and in the case of described radiation source is X-ray tube, advantageously according to described spoke The ratio of the tube current and tube voltage of penetrating source distributes the variance gathering probe value respectively, wherein, false The spectrum of the fixed radiation launched by described X-ray tube is defined by described tube current and described tube voltage.
Described reconstruction unit 10 preferably includes beam hardening correction function, by the spy gathered Measured value applies described beam hardening correction function by the probe value nonlinear mapping that gathered to correction Gather probe value.Described beam hardening correction function is preferably multinomial.Preferably surveyed by calibration Amount or calibrated analog determine described polynomial coefficient, wherein, radiate for by described radiation emission Different spectrum and described area-of-interest in the different absorptions of different component, measure or analog detection Value, and use described probe value to rebuild the image of the described area-of-interest illustrating beam hardening artefact, And wherein, adjust described polynomial described coefficient so that the beam hardening in described reconstruction image Artefact reduces, especially so that described in described reconstruction image, the artefact of beam hardening eliminates.Pin To by the described radiation of described radiation emission different spectrum and with the detection measured respectively or simulate It is worth corresponding difference the first component pad value and second component pad value, can be it is thus determined that different Beam hardening correction function so that beam hardening correction function, institute can be used for gathering probe value State beam hardening correction function and be assigned to corresponding first component pad value and second component pad value, Specifically, the ratio of these respective components pad values, and distribute to by described radiation emission The spectrum of radiation.As above-mentioned it has been mentioned that therefore, if described radiation source is X-ray tube, by The described spectrum of the radiation that described X-ray tube is launched is defined by described tube current and described tube voltage, root There is provided according to corresponding and the first component pad value and second component pad value and described tube current and tube voltage Described beam hardening correction function.
Described reconstruction unit 10 is suitable to by using described beam hardening correction function to repair for beam hardening The probe value just gathered, and be adapted in use to the collection probe value that is corrected together with determined by detect The variance of value, to rebuild the image of described area-of-interest iteratively, specifically, changes according to following For equation:
μ m n + 1 = μ m n + Σ k a k m 1 σ k 2 ( - l o g y k b k - l k n ) Σ k a k m 1 σ k 2 a k , - - - ( 2 )
Wherein, ykIt it is the probe value revised in i & lt is measured, i.e. the measurement density value of correction;bkRepresent The probe value revised when described object does not appears in the described inspection area of described imaging system, as repaiied Positive measurement density value;Represent the image value at intermediate image described in nth iteration;akmRepresent The pictorial element represented by m and the common factor belonging to the ray measuring k;Represent collection probe value respectively Variance;Represent the collection probe value y with correction in nth iterationkThe mould of corresponding absorption value Intend forward projection line integral;And aiExpression is by performing along measuring corresponding ray with i & lt through individual The artificial line integration of the absorption value of the forward projection simulation of body image.
Described iteration is performed until meeting end condition according to equation (2), specifically, until current The difference of intermediate image and intermediate image before less than predefined threshold value.
Refer again to Fig. 1, preferably described iterative approximation image is supplied to display unit 11 with display Described image.
The one-tenth to region of interest domain imaging will be exemplarily illustrated with below according to the flow chart shown in Fig. 2 The embodiment of image space method.In a step 101, it is provided that the collection probe value of described area-of-interest, its In, described area-of-interest includes at least two component, wherein, by being moved relative to transmitting Radiation 4 described radiation source 2 and described area-of-interest and by detect described probe value gather Described probe value, the instruction of described probe value when described radiation source and described area-of-interest relative to each other The radiation after described area-of-interest is passed time mobile.In a step 102, by performing some changing Ride instead of walk and suddenly rebuild the final image of described area-of-interest iteratively, based on institute in described iterative step State and gather probe value and indicate the noise figure of described probe value noise to update intermediate image, and in step In 103, described display unit 11 shows described final image.
Execution during iterative step will be exemplarily illustrated with below according to the flow chart shown in Fig. 3 Step.
In step 201, determine the first component pad value and second component pad value, wherein, described First component pad value is corresponding with the element of the first component in described area-of-interest, and wherein, Described second component pad value is corresponding with the element of the second component in described area-of-interest.In step In rapid 202, for each collection probe value according to corresponding first component pad value, described phase The second component pad value answered and alternatively according to by the spectrum of the described radiation of described radiation emission Determine noise variance.In step 203, described collection probe value is carried out beam hardening correction, and And in step 204, the collection probe value revised and determined by noise variance for updating in Between image, described intermediate image is initial it may be that the most such as, null graph picture.Perform described intermediate image Renewal can be passed through: for generating the artificial line integration of absorption value through described intermediate image forward projection; Determine that the artificial line from the line integral revising the absorption value that probe value determines gathered respectively and absorption value amasss / difference;Calculate based on described difference and described noise variance and update;And institute of rear orientation projection State renewal, specifically, according to equation (2).In iterative step, when described radiation source is about institute State area-of-interest when moving, be preferably carried out described forward projection and institute for each radiation source positions State rear orientation projection, to perform described forward projection and described rear orientation projection at different angle directions, Described probe value is gathered for described each radiation position.In another embodiment, it is possible to use another One known update algorithm updates described intermediate image.
Can one, in several or all iterative steps, or perform to be gathered before iteration starts The correction of probe value.If not performing the beam of the probe value gathered in all of iterative step Hardening correction, for being not carried out the iterative step revised, omits the above step 203.Also can be in institute State before iteration starts, or one, several or all iterative steps perform described noise variance Determination.In the iterative step not determining noise variance, omit step 202.
If described radiation emission heterogeneous x ray, the low energy part of described X-ray compares high-energy Part is absorbed more consumingly.Imaging is had an impact by this effect being referred to as beam hardening." relatively Hard " in beam, the described absorption of tissue is relatively low.In computed tomography image, non-linear Absorption causes so-called beam hardening artefact.Known iterative reconstruction technique is affected by this effect, and And affected in noise characteristic with it by the sensing discordance gathering probe value extraly.Ignore to come From the noise contribution of described detecting devices itself, the described noise that computer tomography is measured come from into Penetrate the statistical Poisson distribution noise of photon.Described detecting devices can perform energy weighting, this meaning Described output signal with
∫dE EN(E), (3)
Proportional, wherein, N (E) is the photon number of ENERGY E.This detector feature is by described noise profile Change into so-called compound Poisson noise, and make described noise energy rely on.The result caused is to penetrate The noise characteristic of probe value is gathered described in bundle hardening effect.Described above to imaging region by using The imaging system of imaging and formation method, it is possible to reduce, especially can eliminate these beam hardenings Impact.
Described imaging system is to revise beam hardening and improve relevant noise model with described formation method, Preferably by beam hardening model combination to described iterative approximation.Can in described iterative reconstruction algorithm Step is revised with one or many application.Described correction step comprises the steps that i) soft from middle Image estimation Organization charts's picture and bone image;Ii) forward projection of these images is performed;Iii) soft tissue estimated is used Absorb and bone resorption estimates that the beam for measurement data (that is, for described collection probe value) is hard Change and relevant correction, and in further iteration, apply this correction to data;And it is iv) logical Soft tissue projection and the bone of crossing approximation project the noise model estimating to improve.
Described imaging system and formation method can apply to dual energy or multi-energy computer tomography Scanning, wherein, the extra image of such as contrast agent figure may be used for improving described beam hardening effect Estimation and for improving the described correction as result.Described correction can be extended to iodine or other by Dual energy or the separate material of multi-energy computer tomography.Described pad value correction and improvement The calculating of noise model can sweep in described calibration steps based on theoretical model or based on calibration steps Retouch particular volume mould with derived parameter correction function.Specifically, above-mentioned formula can be used for base Linear attenuation A in basic materialjPredict described beam hardening effect.Described correction can be based on making apparatus There is the calibration of the body mould such as with bone and soft tissue of various thickness.If scanning these body moulds Various combination, the parameter of simple non-linear beam hardening model, specifically, can estimate multinomial The parameter of formula beam hardening model, and use it for revising.Similar with described beam hardening correction, make an uproar The impact of sound can use theoretical method to estimate or based on calibration measurement.
Although imaging system is computed tomography systems described in the embodiment being described above, but Be described imaging system can also be use probe value rebuild iteratively region of interest area image another become As system.Such as, described imaging system can be x-ray C-arm imaging system or such as monochromatic light Sub-emission computed tomography system or the nuclear imaging system of positron emission tomography system.
Although in the embodiment being described above, described different component is with such as bone and soft tissue not Being correlated with basic material, such as, highly attenuating component and low decay component, described different component also may be used Relate to different physical effect, such as photoelectric effect, Compton effect, K side effect etc..
By studying described accompanying drawing, disclosure and claims, those skilled in the art want in practice It will be appreciated that and realize other modification of the disclosed embodiments when seeking the present invention of protection.
In the claims, word " includes " being not excluded for other elements or step, and indefinite article " one " is not excluded for plural number.
Individual unit or equipment may meet some the functions recorded in the claims.The most not Be recited in mutually different dependent certain measures this only had the fact and is not offered as advantageously making With the combination of these measures.
The determination that performed by one or several unit or equipment, revise, calculating etc., can be by arbitrarily The unit of other numbers or equipment perform.Such as, step 102 and 201 to 204 can be by a list Unit or the different units by arbitrarily other numbers perform.Described determine, revise and calculate, and/or according to Described formation method can be implemented as the journey of computer program and/or specialized hardware to the control of imaging system Sequence code module.
Computer program can provide or as the portion of other hardware using stored/distributed together with other hardware On the suitable medium divided, such as optical storage media or solid state medium, it is possible to issue otherwise, example As, by the Internet or other wired or wireless communication systems.
Any reference in claim is all not to be read as restriction scope.
The noise of the probe value gathered by imaging system can depend on being imaged in the region of interest The contribution of different component, described different component is detected through with the collection produced respectively by described radiation Value.This dependence is considered, wherein, really when rebuilding the image of described area-of-interest iteratively The fixed first component pad value corresponding with the element of the first component in described area-of-interest, and The second component pad value corresponding with the element of the second component in described area-of-interest, wherein, Determine noise figure from described first component pad value and described second component pad value, and wherein, make Described image is updated with described noise figure.Consider to gather described in probe value on described different component The dependence of noise improves the quality of described iterative approximation image.

Claims (11)

1., for the imaging system to region of interest domain imaging, described imaging system (30) including:
-probe value provides unit (31), and it is for providing the probe value of the collection of described area-of-interest, Wherein, described area-of-interest includes at least two component, and wherein, described probe value is by following Gathered: be moved relative to launch the radiation source (2) of radiation (4) and described region of interest Territory and detect instruction and wear when being moved relative to when described radiation source (2) and described area-of-interest Cross the described probe value of described radiation (4) after described area-of-interest;
-reconstruction unit (10), it is emerging for rebuilding described sense iteratively by the some iterative steps of execution The final image in interest region, probe value based on described collection and instruction in described some iterative steps The noise figure of the noise of described probe value updates intermediate image, and wherein, described reconstruction unit (10) is fitted In:
-determine the first component pad value and second component pad value, wherein, described first component pad value Corresponding with the element of the first component in described area-of-interest, and wherein, described second component Pad value is corresponding with the element of the second component in described area-of-interest;
-determine described noise figure from described first component pad value and described second component pad value,
Wherein, described reconstruction unit (10) is suitable to determine corresponding for the probe value of each collection Ratio between one component pad value and corresponding second component pad value, and be suitable to according to described ratio Rate determines described noise figure.
2. imaging system as claimed in claim 1, wherein, described reconstruction unit (10) is suitable to base First component pad value determined by and determined by second component pad value for beam hardening correction The probe value of described collection.
3. imaging system as claimed in claim 1, wherein, described noise figure is described probe value Variance.
4. imaging system as claimed in claim 1, wherein, sends out according to by described radiation source (2) The energy spectrum of the described radiation (4) penetrated determines described noise figure.
5. imaging system as claimed in claim 1, wherein, described reconstruction unit (10) is suitable to lead to Cross described first component pad value identified below and described second component pad value:
-determine the first component Parts and second component part, wherein, described first from described intermediate image Component Parts illustrates the element of described first component in described area-of-interest, and wherein, described The element of the described second component in second component described area-of-interest shown partially;
-determine that described first component decays by simulation by the forward projection of described first component Parts Value, and determine that described second component declines by simulation by the forward projection of described second component part Depreciation.
6. imaging system as claimed in claim 5, wherein, described first component Parts forms first Constitutional diagram picture, and described second component part formation second component image, and wherein, described heavy Build unit (10) to be suitable to determine described the by simulation by the forward projection of described first constitutional diagram picture One component pad value, and determined described by the forward projection of described second component image by simulation Second component pad value.
7. imaging system as claimed in claim 1, wherein, described probe value provides unit (31) It is adapted to provide for the probe value of the collection of Energy Dependence, and wherein, described reconstruction unit (10) is suitable to By the probe value application foundation decomposition of components method of the collection of described Energy Dependence is determined described first Component pad value and described second component pad value.
8. imaging system as claimed in claim 7, wherein, the detection of the collection of described Energy Dependence It is worth corresponding with multiple energy branch mailbox, and wherein, described reconstruction unit (10) is suitable to by using The probe value of collection is described as the probe value of the described collection of the combination of the first contribution and the second contribution The simultaneous equations of the probe value of the collection for multiple Energy Dependences are solved by model, apply described Basic components is decomposed, and the first Energy Dependence caused by described first component is depended in described first contribution Decay, the decay of the second Energy Dependence caused by described second component is depended in described second contribution.
9. imaging system as claimed in claim 1, wherein, described first component pad value and bone unit Element is corresponding, and described second component pad value is corresponding with soft tissue element.
10., for the formation method to region of interest domain imaging, described formation method includes:
-rebuild the final image of described area-of-interest iteratively, in institute by performing some iterative steps State the probe value of collection based on described area-of-interest and the described probe value of instruction in some iterative steps Noise noise figure update intermediate image, wherein, described area-of-interest includes at least two component, Wherein, described probe value is collected by following: be moved relative to launch the radiation of radiation Source and described area-of-interest and detect instruction when described radiation source and described area-of-interest relative to Through the described probe value of the radiation after described area-of-interest when moving each other, wherein it is determined that the One component pad value and second component pad value, wherein, described first component pad value is emerging with described sense The element of the first component in interest region is corresponding, and wherein, described second component pad value and institute The element stating second component in area-of-interest is corresponding;Wherein, from described first component pad value Determining described noise figure with described second component pad value, wherein, the probe value for each collection is true Ratio between fixed corresponding first component pad value and corresponding second component pad value, and be suitable to Described noise figure is determined according to described ratio.
11. 1 kinds of imaging devices being used for region of interest domain imaging, described imaging device includes:
-for rebuilding the final image of described area-of-interest iteratively by performing some iterative steps, The probe value of collection based on described area-of-interest and the described spy of instruction in described some iterative steps The noise figure of the noise of measured value updates the unit of intermediate image, wherein, described area-of-interest include to Few two components, wherein, described probe value is collected by following: be moved relative to send out Penetrate the radiation source of radiation and described area-of-interest and detect instruction when described radiation source and described sense emerging Interest region passes the described probe value of the radiation after described area-of-interest when being moved relative to, Wherein it is determined that the first component pad value and second component pad value, wherein, described first component decay It is worth corresponding with the element of the first component in described area-of-interest, and wherein, described second group Divide pad value corresponding with the element of the second component in described area-of-interest;Wherein, from described One component pad value and described second component pad value determine described noise figure, wherein, adopt for each The probe value of collection determines the ratio between corresponding first component pad value and corresponding second component pad value Rate, and be suitable to determine described noise figure according to described ratio.
CN201180060794.5A 2010-12-17 2011-12-15 For the imaging system to region of interest domain imaging Active CN103262124B (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
EP10195738 2010-12-17
EP10195738.9 2010-12-17
PCT/IB2011/055689 WO2012080971A1 (en) 2010-12-17 2011-12-15 Imaging system for imaging a region of interest

Publications (2)

Publication Number Publication Date
CN103262124A CN103262124A (en) 2013-08-21
CN103262124B true CN103262124B (en) 2016-11-09

Family

ID=45470619

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201180060794.5A Active CN103262124B (en) 2010-12-17 2011-12-15 For the imaging system to region of interest domain imaging

Country Status (6)

Country Link
US (1) US9230348B2 (en)
EP (1) EP2652709B1 (en)
CN (1) CN103262124B (en)
BR (1) BR112013014802A2 (en)
RU (1) RU2013132960A (en)
WO (1) WO2012080971A1 (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012095713A1 (en) * 2011-01-10 2012-07-19 Koninklijke Philips Electronics N.V. Dual -energy tomographic imaging system
US10685452B2 (en) * 2015-06-30 2020-06-16 Koninklijke Philips N.V. Registration apparatus for registering images
US20170039735A1 (en) * 2015-08-06 2017-02-09 General Electric Company Computed tomography self-calibration without calibration targets
EP3267893B1 (en) * 2016-01-29 2018-08-15 Koninklijke Philips N.V. Cone beam computed tomography projection values providing system and method
JP6923562B2 (en) 2016-02-19 2021-08-18 エス. カリム、カリム X-ray detection system and X-ray detection method
US11353533B2 (en) 2016-02-24 2022-06-07 Ohio State Innovation Foundation Methods and devices for contrast agent magnetic resonance imaging
JP7063378B2 (en) * 2018-03-20 2022-05-09 日本電気株式会社 Imaging device and imaging method
EP3683765A1 (en) * 2019-01-18 2020-07-22 Koninklijke Philips N.V. System for determining a tissue-specific property

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101320475A (en) * 2008-06-10 2008-12-10 北京航空航天大学 Operating range estimation method of infrared imaging system under complex background condition

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6507633B1 (en) * 2001-02-15 2003-01-14 The Regents Of The University Of Michigan Method for statistically reconstructing a polyenergetic X-ray computed tomography image and image reconstructor apparatus utilizing the method
JP4854137B2 (en) 2001-06-21 2012-01-18 株式会社東芝 Medical diagnostic imaging equipment
US6754298B2 (en) * 2002-02-20 2004-06-22 The Regents Of The University Of Michigan Method for statistically reconstructing images from a plurality of transmission measurements having energy diversity and image reconstructor apparatus utilizing the method
US6775352B2 (en) 2002-08-16 2004-08-10 Ge Medical Systems Global Technology Company, Llc Method and system for implementing variable x-ray intensity modulation schemes for imaging systems
US6990171B2 (en) * 2003-10-27 2006-01-24 General Electric Company System and method of determining a user-defined region-of-interest of an imaging subject for x-ray flux management control
JP4709162B2 (en) 2003-12-16 2011-06-22 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ Correcting artifacts caused by the heel effect
US7391844B2 (en) 2005-01-14 2008-06-24 General Electric Company Method and apparatus for correcting for beam hardening in CT images
US7747057B2 (en) 2006-05-26 2010-06-29 General Electric Company Methods and apparatus for BIS correction
US8005288B2 (en) * 2007-04-24 2011-08-23 Siemens Aktiengesellschaft Layer reconstruction from dual-energy image pairs
DE102009015773A1 (en) * 2008-04-10 2010-02-04 Siemens Aktiengesellschaft Method for processing reconstructed CT image data sets and CT system
US8194961B2 (en) * 2008-04-21 2012-06-05 Kabushiki Kaisha Toshiba Method, apparatus, and computer-readable medium for pre-reconstruction decomposition and calibration in dual energy computed tomography

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101320475A (en) * 2008-06-10 2008-12-10 北京航空航天大学 Operating range estimation method of infrared imaging system under complex background condition

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Noise considerations in dual energy CT scanning;Frederick Kelcz等;《Medical Physics》;19790930;第6卷(第5期);第418-425页 *

Also Published As

Publication number Publication date
EP2652709B1 (en) 2015-06-17
WO2012080971A1 (en) 2012-06-21
RU2013132960A (en) 2015-01-27
US20130279785A1 (en) 2013-10-24
US9230348B2 (en) 2016-01-05
CN103262124A (en) 2013-08-21
BR112013014802A2 (en) 2019-09-24
EP2652709A1 (en) 2013-10-23

Similar Documents

Publication Publication Date Title
CN103262124B (en) For the imaging system to region of interest domain imaging
CN106233335B (en) X-ray spectrum imaging method and system
CN104024886B (en) The detection device of photon is detected for considering pile-up events
EP3050028B1 (en) Joint reconstruction of electron density images.
US10679385B1 (en) System and method for statistical iterative reconstruction and material decomposition
CN103299345B (en) Dual energy tomographic imaging system
CN103339497B (en) Probe value treatment facility
CN102483853B (en) For the treatment of the apparatus and method of data for projection
JP2010510514A (en) Apparatus and method for determining a detector energy weight function of a detection unit
US10420519B2 (en) Computed tomography system
CN107635469A (en) The estimation of the decay pattern met based on the scattering in PET system
CN101547648A (en) CT imaging system
Shi et al. Library based x‐ray scatter correction for dedicated cone beam breast CT
Kengyelics et al. X-ray system simulation software tools for radiology and radiography education
US20150317777A1 (en) Image processing apparatus and method for filtering an image
CN106471393B (en) Spectrum material for photon counting application decomposes
US11241211B2 (en) Method and apparatus for singles spectrum estimation and for dead-time correction in positron emission tomography (PET)
Colijn et al. Experimental validation of a rapid Monte Carlo based micro-CT simulator
CN103620393B (en) Imaging device
Kaplan et al. A differential attenuation method for simultaneous estimation of SPECT activity and attenuation distributions
US7385200B2 (en) Re-binning method for nuclear medicine imaging devices
CN106133792A (en) Video generation device
Yao et al. An analytical approach to estimating the first order scatter in heterogeneous medium. II. A practical application
Pan et al. A transmission measurement-based spectrum estimation method incorporating X-ray tube voltage fluctuation
Vijayalakshmi et al. Comparison of algebraic reconstruction methods in computed tomography

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant